Gino, for quite a long time now you have been trying to get at the truth of audio engineering by asking lots of question about all sorts of issues and then being confused when the answers given to you by different people are different or even contradictory. Sadly, that is the state of audio - lots of opinions but not enough engineering. You are seeking answers, when you should be seeking understanding. Given some understanding, you would be able to decide for yourself which answers you believe - you would even be able to answer some questions posed by others.
Seek understanding. In many cases this will involve numbers.
Seek understanding. In many cases this will involve numbers.
It is engineering, it usually involves numbers, and digital audio kind of involves numbers by definition.....
One thing to get a handle on early (which many in the audio game never really do) is the effect of orders of magnitude, by which I mean that there are plenty of effects which while real, can be ignored most of the time because they are tiny compared to the elephant in the room (whatever that happens to be).
A good example would be something like skin effect in speaker leads, yea, akin effect is real and you could even (maybe) measure it in a sufficiently low Z speaker circuit, but that does NOT make it something to care about because it is at most going to be small fractions of a dB and moving your head 6 inches will have far more effect (Now in a switched mode power supply magnetic on the other hand, it might be the difference between works and smokes)....
There are all kinds of odd corners of physics of this nature that get wheeled out by the hard of thinking as being things to worry about, run the numbers and they usually get moved to the unimportant category (And when they don't, you have learned something).
As I say, lots of fashion at the high end, and pointing out the emperors sartorial challenges does not go down well.
Regards, Dan.
One thing to get a handle on early (which many in the audio game never really do) is the effect of orders of magnitude, by which I mean that there are plenty of effects which while real, can be ignored most of the time because they are tiny compared to the elephant in the room (whatever that happens to be).
A good example would be something like skin effect in speaker leads, yea, akin effect is real and you could even (maybe) measure it in a sufficiently low Z speaker circuit, but that does NOT make it something to care about because it is at most going to be small fractions of a dB and moving your head 6 inches will have far more effect (Now in a switched mode power supply magnetic on the other hand, it might be the difference between works and smokes)....
There are all kinds of odd corners of physics of this nature that get wheeled out by the hard of thinking as being things to worry about, run the numbers and they usually get moved to the unimportant category (And when they don't, you have learned something).
As I say, lots of fashion at the high end, and pointing out the emperors sartorial challenges does not go down well.
Regards, Dan.
Hi and thanks a lot for having pointed out the problem.
Lack of technical education. The good sense makes up to a point.
A lot could be learned looking inside units and above all at their schematic i think. Parts used in the best reviewed units.
And this is not an easy thing to do. And a scope ... i need a scope.
Thanks a lot again. Gino
Schematics (and board layouts), yes to a point.
Which parts, generally not so much, as that gets into complicated tradeoffs with the schematics, BOM cost, marketing buzzwords.... I don't say it does not matter, but what matters is picking the right topology and then the right parts for that topology.
A scope is good, plenty of not overly expensive second hand ones out there.
Regards, Dan.
Which parts, generally not so much, as that gets into complicated tradeoffs with the schematics, BOM cost, marketing buzzwords.... I don't say it does not matter, but what matters is picking the right topology and then the right parts for that topology.
A scope is good, plenty of not overly expensive second hand ones out there.
Regards, Dan.
Gino, I am so happy for you, that you are asking questions. You need to investigate and decide for yourself. I can give you a bottom line wherever you like, based on what I understand with different types of amplifiers and the experiences I can translate to individuals needs.
However, I'm having trouble understanding you. You now are switching topics, frequently and without explaining your train of thought. I say this because after your first question, you then moved on to a small signal generator in a DAC. But then you switched to a clock generator. And then a multiple redundant oscillator setup. Mechanically speaking, this is like going from a car, to a scooter, to an e-bike, to a tricycle. They all have their own unique challenges.
No one can help you answer all questions in the same answer in multiple states of change.
You want to design something but from our discussions so far, don't know what you are looking at. A scope will not tell you anything if you cannot interpret the meaning, just like a speedometer is useless unless you are measuring barometric (Hg) pressure.
The best place, is to start at the beginning. We can give you our experiences, our understanding but first hand knowledge and empirically based deduction is key.
If you want a couple of examples of some interesting ways to account for some design considerations, it is best to start simply.
If you want something complex and nuanced, read on.
There are a number of good examples of DAC engineering with an open approach as to reasoning and implementation considering all parts, including the power supply considerations.
DDDAC, I have seen, spent a great amount of time working with a very popular DAC chip that uses a dmos output stage, even to the point of stacking multiple dac's in parallel to get some nice results. These units are absolutely rely on consistent timing and execution of signal reproduction in unison. So if you are wanting to learn about that project, here is the link:DDDAC 1794 NOS DAC - Non Oversampling DAC with PCM1794 - no digital filter - modular design DIY DAC for high resolution audio 192/24 192kHz 24bit
Bottom line, you have more questions with more variables than I think you are aware of. For example, your note about low level "frying pan noise" could have more to do with other parts of the system that has nothing to do with power supply variation. Maybe, rather than working to ideate your "Magnum Opus" on the first try, maybe start with actually building something that works, and build your knowledge from that first project.
Good idea getting a scope. Valuable observation tool for many uses.
I will leave you with this. There are many answers. The question is which one meets your needs.
You must define your needs. Some people call these specifications. Once you design to the specifications, then perhaps you will be happy with the results. I think Sy has said something very close to that. The alternative is to design without regard to the specification (needs) and accept the results - which are usually disappointing.
Best of luck Gino.
However, I'm having trouble understanding you. You now are switching topics, frequently and without explaining your train of thought. I say this because after your first question, you then moved on to a small signal generator in a DAC. But then you switched to a clock generator. And then a multiple redundant oscillator setup. Mechanically speaking, this is like going from a car, to a scooter, to an e-bike, to a tricycle. They all have their own unique challenges.
No one can help you answer all questions in the same answer in multiple states of change.
You want to design something but from our discussions so far, don't know what you are looking at. A scope will not tell you anything if you cannot interpret the meaning, just like a speedometer is useless unless you are measuring barometric (Hg) pressure.
The best place, is to start at the beginning. We can give you our experiences, our understanding but first hand knowledge and empirically based deduction is key.
If you want a couple of examples of some interesting ways to account for some design considerations, it is best to start simply.
If you want something complex and nuanced, read on.
There are a number of good examples of DAC engineering with an open approach as to reasoning and implementation considering all parts, including the power supply considerations.
DDDAC, I have seen, spent a great amount of time working with a very popular DAC chip that uses a dmos output stage, even to the point of stacking multiple dac's in parallel to get some nice results. These units are absolutely rely on consistent timing and execution of signal reproduction in unison. So if you are wanting to learn about that project, here is the link:DDDAC 1794 NOS DAC - Non Oversampling DAC with PCM1794 - no digital filter - modular design DIY DAC for high resolution audio 192/24 192kHz 24bit
Bottom line, you have more questions with more variables than I think you are aware of. For example, your note about low level "frying pan noise" could have more to do with other parts of the system that has nothing to do with power supply variation. Maybe, rather than working to ideate your "Magnum Opus" on the first try, maybe start with actually building something that works, and build your knowledge from that first project.
Good idea getting a scope. Valuable observation tool for many uses.
I will leave you with this. There are many answers. The question is which one meets your needs.
You must define your needs. Some people call these specifications. Once you design to the specifications, then perhaps you will be happy with the results. I think Sy has said something very close to that. The alternative is to design without regard to the specification (needs) and accept the results - which are usually disappointing.
Best of luck Gino.
I'll add one more comment. "best reviewed units". Reviewed by who? Most published reviews are more an evaluation of how much the reviewer gets out of doing the review and has nothing actually to do with the equipment being reviewed.
I believe in listening test ONLY when the listener does not know what is being listened to, and levels are matched. You can talk yourself into pretty much anything...
I believe in listening test ONLY when the listener does not know what is being listened to, and levels are matched. You can talk yourself into pretty much anything...
Gino, as a simple solution what I do is download the video, there are plenty of software solutions out there for doing this, I use YouTube Downloader HD. I'm only interested in the audio component, so I then use Audacity software to strip this out, then simply export as a WAV file. At times I have upsampled to hi res formats, which has improved the "musicality" on at least one machine. Then, use the 'best' media player to replay, at the moment for me this is Media Monkey.
Obviously very longwinded and a chore, only worth it if there is a special desire or need to get the playback as good as possible, 🙂.
Hi and thanks a lot again for the very kind and helpful support.
I understand completely how complex is the task and usual i start from the end and go backwards. 😱
You are right i am confused. Let me resume the problem.
I have decided to build my AV system using a pc as a source.
I have some dacs at hand to try, some should be ok other less ok.
Anyway i need a way to connect these dacs to the pc.
I found and bought a nice usb to spdif converter that works ok.
I am using its AES/EBU out to the dacs.
Then i start reading, from companies selling custom power supplies, how important is power supply quality with digital equipment.
So i look inside this converter and at the same time inside the best similar products and i saw differences in the power supply, like it should given the price difference.
My unit has no filtering on the mains and cheap diode bridges.
So the idea was to replace the transformer with one with better filtering effect and diodes with Schottky and ps caps.
Just this. It should do not harm i guess.
I have no possibility with the pcb. If there are design flaws there they will remain for sure. And it could be the case.
very true. I agree. It is a difficult instrument.
Thanks a lot but i see much more challenging that just upgrading a ps.
This is just what i would like to do and learn as a first.
I am very intrigued by power supplies and their issues.
Thanks again for the very valuable advice.
Just to put the money where my mouth is i have decided for a buy from Mouser.
I have seen a Hammond mains transformer of the type used in my reference unit (see picture below) and bough it.
An externally hosted image should be here but it was not working when we last tested it.
Then i have bought some Schottky diodes and some Nichicon FG caps to replace the stock ones.
Would this be beneficial ? i hope so.
The new transformer with splitted bobbins should be better at blocking noise from the mains. Schottky diodes should generate less noise than the poor stock ones and the caps should also be beneficial.
If the real flaws are instead in the pcd design i will not able to do anything, of course.
This is just a try. Main problem is the space inside, very limited.
A real shame.
But i could keep the new ps outside ... i have to see.
Thanks a lot again.
Kind regards, gino
Hi of course you are right. How dependable are the reviewers.
But when a unit gets praise from all who have tested it ... well praise is too little. Words like "... redefines what we can expect from digital ... " sound more strong that just a compliment. This is an achievement.
As i said i have decided to focus my attention on low V/low VA power supplies.
So i will be reading a lot the relevant section and try to build something.
I think that a really great unit starts with a very well designed and built power supply.
It is just a start i agree. But a fundamental step.
Honestly with all the project using smd parts these days i see no chance for me.
I have to start from an already populated pcb ... no chance at all.
I have a very strong feeling that this aspect of the power supply is not very well treated in cheap units.
When i look at the ps of my converter and the one on the reference converter i see many differences.
I will report anyway. It will take time because the parts are not in stock.
By the way i am impressed by Mouser catalog. It is an extremely educative web site. Each part comes with its data sheet.
This is a very very good site indeed. I already love it and i think i will buy from there a lot in the next years. They have also a strict policy against fale parts.
Thanks a lot again.
Kind regards, gino
Last edited:
Thanks for the valuable advice.
Problem is that i am interested also in the video.
Actually it is indeed concerts with video that i like best. There is no comparison with just listening to music.
I have seen some concerts on blue ray and they are just sublime.
I send the video to the tv and listen audio from my stereo system through usb > converter > dac > stereo rig.
So i am using the tv as a monitor only. No sound from it.
I have an extremely bad opinion of the hdmi connector vs. the DVI (this one is magnificient instead).
I do not understand why they have not used DVI and also for carrying audio digital signals.
Thanks a lot again for your helpful advice.
Regards, gino
Last edited:
Gino,
Some interesting ideas.
Have you tried measuring what mains noise there is?
I recently traced a large source of mains noise in my home to a wired ethernet router power supply. It did not have any back EMF protection, so all the reflected garbage signals from the ethernet wires was being dumped into my power lines instead of being filtered out.
As a result, I couldn't get any AM radio signal inside my home, it was all garbage. I spent weeks trying to filter out out from my radio receiving unit but never succeeded. That was, until I traced the source down to my ethernet. All I had to do was put a proper SMPS on the ethernet switch instead of the small switching power supply brick that came with it. The switch was different than the router from the phone company this was for distribution of ethernet to several other wired devices in my home.
Some interesting ideas.
Have you tried measuring what mains noise there is?
I recently traced a large source of mains noise in my home to a wired ethernet router power supply. It did not have any back EMF protection, so all the reflected garbage signals from the ethernet wires was being dumped into my power lines instead of being filtered out.
As a result, I couldn't get any AM radio signal inside my home, it was all garbage. I spent weeks trying to filter out out from my radio receiving unit but never succeeded. That was, until I traced the source down to my ethernet. All I had to do was put a proper SMPS on the ethernet switch instead of the small switching power supply brick that came with it. The switch was different than the router from the phone company this was for distribution of ethernet to several other wired devices in my home.
Hi obviously not. How can i do that ?
I have found now at home a scope i bought and never used, a Hantek 6022BE with usb port.
I do not have a clue how to use it by the way. But i will try to learn.
Maybe it could be useful to "see" noise in circuits ?
This is my main inspection tool now.
http://www.amazon.com/Hantek-Digital-Storage-Oscilloscope-Bandwidth/dp/B009H4AYII
I wonder how you managed to trace the source of noise.
In my case i was thinking to put together again the device as stock, more or less and look for the noise along the circuit (but i have to learn how to do that). In order to assess which kind of noise is.
Maybe it will be at a tolerable level.
I have to try this thing i have at hand now. Maybe it will convince me that i am just hearing things.
Thanks again, gino
Good luck.
It was simpler than you think. I just started unplugging things in my house and went back to the speaker to see if the noise was still there after I unplugged each device, one by one. Once I made the noise go away, I just added back individual devices until the sound re-appeared.
No scope, no DMM just process of elimination and ears.
It was simpler than you think. I just started unplugging things in my house and went back to the speaker to see if the noise was still there after I unplugged each device, one by one. Once I made the noise go away, I just added back individual devices until the sound re-appeared.
No scope, no DMM just process of elimination and ears.
Thanks for the very interesting hint ! Very practical way actually.
It is true that noise is around the corner. I had some noise in the speakers
then i disconnected the dvd player and the noise disappeared.
I think it was a ground loop i guess.
Thanks again. gino
Hi,
I have done a lot of tests and measurments on the ac lines. Most unwanted noise freqs comes from within the home... 99%. 1% from outside the home. SMPS and computers, lamp dimmers, TV, anything digital and on and on. Each item needs to be filtered, individually or isolated from each other and from the line.
THx-RNMarsh
The unplugging of each device in the house technique can easily be extended to beyond just hearing if there is an obvious noise from the speaker - if the sound is not 'right', there's an unpleasant, harsh, flat, uninteresting, or whatever, quality to it, try the same procedure - with luck you might find some prime offenders ... and then work out a strategy of dealing with that ...
But why not ? For just a few more bucks , <10uV noise and -125db PSRR is
a layout away.
Do we all have to go down the road of cheap Ebay or OEM offerings ?
In the end , noise and other supply "demons" can be eliminated , freeing up
the designer to create better audio concentrating on the other parts of the circuit.
Same with the Linear stuff , use capacitance multipliers and other "enhancements" ,
even if there is no real problem , If you do have a circuit that is PSRR "challenged" ,
it won't really matter. DIY should be overdesigned/overbuilt. Short of this ,
just go buy some OEM/Walmart junk.
OS
The only way to answer the "Who to trust?" and "Who is right?" questions is to get enough knowledge to answer the "Is this true?" questions yourself.
Hi and thanks a lot for the very helpful reply.
It depends on what are you looking for.
If your passion is electronics i see the excitement in the challenge to design something high end in terms of measurements.
In the end i think that for me electronics is just a way to get music.
And money is of course an issue.
this sounds very reasonable. Noise is noise not music.
And i like very low noise units simply because i get more music from the recordings played with these units.
I agree. Anyway i still looking for a way to get very clean 9-10 VDC to supply a unit i have ... current draw should be not more the 0.2-0.3 A but i do not know the actual value.
Looking at its power supply and then at the power supplies of similar but better performing units i saw important differences.
I would just need a very good kit (not smd) for a very low noise regulator.
I have already suitable transformers.
Another solution would be to provide unregulated DC to the circuit that has already a lt1963 on board.
Clearly i would prefer to feed it with a very clean DC.
This is a very very important unit for me.
I tried it with different pc from extremely cheap to more expensive and always has satisfied me.
But i would like just one drop more of juice ... the lower the noise the better, ever. I hate noise.
Art of noise aside ...
Thanks again, gino
https://www.youtube.com/watch?v=pDvm_MlgxdU
- Status
- Not open for further replies.